EP3818478A1 - Procédé utilisant des réseaux de neurones artificiels pour trouver un code de système harmonisé unique à partir de textes donnés et système pour le mettre en ?uvre - Google Patents

Procédé utilisant des réseaux de neurones artificiels pour trouver un code de système harmonisé unique à partir de textes donnés et système pour le mettre en ?uvre

Info

Publication number
EP3818478A1
EP3818478A1 EP18849466.0A EP18849466A EP3818478A1 EP 3818478 A1 EP3818478 A1 EP 3818478A1 EP 18849466 A EP18849466 A EP 18849466A EP 3818478 A1 EP3818478 A1 EP 3818478A1
Authority
EP
European Patent Office
Prior art keywords
unique identification
merchandise
items
artificial neural
identification numbers
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP18849466.0A
Other languages
German (de)
English (en)
Inventor
Asim BARLIN
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Solmaz Gumruk Musavirligi AS
Original Assignee
Solmaz Gumruk Musavirligi AS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Solmaz Gumruk Musavirligi AS filed Critical Solmaz Gumruk Musavirligi AS
Publication of EP3818478A1 publication Critical patent/EP3818478A1/fr
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/018—Certifying business or products
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3347—Query execution using vector based model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/284—Lexical analysis, e.g. tokenisation or collocates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services

Definitions

  • the invention disclosed hereby generally concerns a method for allocating specific/unique numbers for any suggested item according to an established group of standards such as the Harmonized System (HS) which is the internationally valid generalized commodity nomenclature standard, and a system implementing said method for designating said specific/unique identification numbers to any suggested item or merchandise.
  • HS Harmonized System
  • the Harmonized Commodity Description and Coding System generally referred to as "Harmonized System” or simply "HS” is a multipurpose international product nomenclature developed by the World Customs Organization (WCO).
  • WCO World Customs Organization
  • HS Harmonized Commodity Description and Coding System
  • WCO World Customs Organization
  • a considerable amount of susceptibility to human error is present in the technique at this stage, magnitude of data processing load stemming from the great body of commodity groups (approx. 5000) having been arranged in a logical and legal structure notwithstanding.
  • To address these concerns present invention weighs on a supervised learning-based method in means of technical rigor; i.e. a decision mechanism based on a neural network having been trained with data of five clusters that is capable of confident conceptual classification.
  • the document denoted with the publication number KR 101571041 ( B1 ) discloses a system for harmonized system (HS) classification of merchandise with an interface processing unit to select an interface for input reception; a database having HS code correspondence information; and an HS code determination unit. Said disclosure is subject to a set of systemwise rules and constraints such as inclusion/exclusion, and requires explicit user input.
  • US 2016275446 (A1 ) discloses an apparatus and a method for determining HS code, receiving selectable determination factors from a user and ascertains similarity through comparison with memory storage.
  • WO 201 6057000 (A1 ), identified as one of the publications in the present technical field, defines a method of learning-oriented classification stating ad hoc and learning rules subsequently employed in order to increase the overall power of handling for a wide array of goods.
  • Other prior art documents such as E P 2704066 (A1 ) relates to classification of transactions based upon analysis of multiple variables which is deemed to facilitate identification and categorization of an item.
  • US 20050222883 (A1 ) discloses a general system and method for brokerage operations support that follows steps including receiving information pertaining to the shipment rates of a specific country and is characterized by an architecture encompassing a server and different workstations, attended and unattended alike.
  • Obj ects of the P resent I nvention Primary object of the present invention is to provide a method and a system for allocating customs tariff numbers according to Harmonized System (HS) to item and merchandise that is characterized by a processing unit specialized for an artificial neural network- based supervised learning application trained with 5 clusters of conceptualized categories, in turn posing a strong and more accurate, as well as more precise alternative when regarded next to the prior art.
  • HS Harmonized System
  • an item classification and unique identifier allocation scheme is implemented that centers mainly around artificial neural networks, a supervised learning algorithm; however user intervention and need for explicit input are minimized while reducing errors and improving integrity.
  • a processing unit works in harmony with a database and an input means for an item subject to trade to be verified in customs, as opposed to the standard method of manual inspection generally taking precedence.
  • Machine learning method as the main part of the disclosed invention is inspired by and modeled after aforementioned standard method of manual inspection carried out by actual human beings, inconclusiveness of which leads to a binding report which takes time to issue and process.
  • a query is formed as a result of a series of preprocessing steps.
  • Text is made all lowercase and parsed, subsequently converted to all UTF8 characters, after which they are broken down to syllables and reorganized in the form of phonemes and morphemes, and later into word groups of two and three.
  • the vector conversion of a combination of all said phonemes, morphemes and word groups takes place, which is subsequently normalized.
  • the training of the artificial neural network which is combined with an output layer function in order to minimize the error and reduce over/underfitting, once which is achieved the model is saved as a file on the local disk.
  • Models to be trained according to HS-based data are designed according to previously mentioned grouping of numbers resulting in the creation of different divisions of training. Consequently, the output of category suggestion neural network is fed to the chapter suggestion/prediction neural network, the output of which is, in turn, fed to position suggestion/prediction neural network and so on. Final set of suggestions/predictions are then displayed on screen to a human user.
  • Fig. 1 demonstrates a possible relationship diagram concerning the customs number allocation method according to the present invention.
  • Fig. 2 demonstrates the preprocessing steps executed before the training phase according to the present invention.
  • Fig. 3 demonstrates the file hierarchy of the training sets concerning the artificial neural network for hypothetical items according to the present invention.
  • Fig. 4 demonstrates an exemplary diagram for showing a purported flow of information regarding different artificial neural networks when suggesting a number to an item according to the present invention.
  • Fig. 5 demonstrates the significant preprocessing steps and training processes of artificial neural networks on every level according to the present invention.
  • Fig. 6 demonstrates the entire artificial intelligence model that includes preprocessing, training and prediction/suggestion as a whole according to the present invention.
  • the present invention discloses a method and a system implementing the same method for automatically allocating a unique number to a trade item or merchandise based on a query formed.
  • Artificial Neural Networks are used as the key element of the method in order to enable enhanced flexibility to an otherwise strict and convoluted process, as well as a reduced error for a broadly more straightforward approach.
  • a unique number to be ascribed to an item in question may have different components in numeric format, with different groups of which carry information pertaining to a different aspect or feature of said item. Such a process has numerous factors to be considered and many properties as well as constraints to be scrutinized; making it laborious and time- consuming.
  • disclosed method and system comprise features of data processing significantly arranged to utilize machine learning concepts on processing mechanisms handling a wide array of information, replacing actual human beings executing the tasks aforementioned.
  • Different items pose different properties and use areas which become matters of consideration when assigning unique identifiers to them, often requiring voluminous look-up tables ( LUTs) and incidentally laboratory inspections.
  • Disclosed invention addresses these issues all the while proposing a systematic and programmatic number/identifier allocation scheme that makes use of the conceptual classification property applicable to systems and problems in question.
  • Direct input is the manual text entry, which relates to the verbal definition of an item/merchandise in the desired language; whereas indirect inputs refer to catalogue(s) and technical specifications used in the assessment of said item/merchandise externally.
  • Direct input arrives raw, therefore needs to be processed prior to be used for text search and training purposes according to a custom guideline. Once text search is concluded and output is obtained, legislation and item databases begin to deliver according to the furthering of assessment procedures.
  • Legislations database may include, in multiple languages, informations referring to the country/state, relevant fiscal info such as tax and funds, legislation pertaining to liabilities and sanctions, if present, and comparative legal status between two countries/states or similar items/merchandise of the user s choice.
  • Output layer function delivers suggestion result to the final state of the output layer, which is utilized for both training and classification instances.
  • the function, where z represents suggestion result, /the index of the suggested label, and y the total number of labels is given below:
  • Disclosed invention differs from other full text search and machine learning methods in the sense that it categorizes the classification of subject matter into conceptual constituents, therefore mimicking the implicit decision mechanism taking place in real time significantly better: Same data are processed in batches of five, marginalizing the error at every level via ensuring quadruple error rate dissolution between thereof, as opposed to the amount of epochs needed for training to bring error rate down to a desired level. In doing so, indirect conceptualization of data is assured, akin to that of an actual human being compartmentalizing said unique identification number into smaller parts to progressively inspect said item/merchandise.
  • Post-training system initially executes at level one, during which a category for the item/merchandise is suggested. Suggested category is then prepared as the input of the next level, which is level two, output of which yields a suggestion for chapter and delivered to serve as input to the next level, level 3, and so on.
  • level two the next level
  • level 3 the next level
  • the end outcome is presented in a form such that it may include close, overlapping instances of unique identification numbers, in cooperation with the other modules of the system such as legislation database and parallel item/merchandise lists, according to at least one embodiment. In this, it is legitimate to reflect that this system and method do not strictly confine the operation to the exact classifying behavior of an actual human being, i.e.
  • usage of said method and system are facilitated further via integration of various other components such as a legislation database, and item database, a ' similar items module, all of which serve to extend the practical grasp of disclosed invention.
  • a legislation database and item database
  • a ' similar items module all of which serve to extend the practical grasp of disclosed invention.
  • One embodiment is, for instance, capable of requesting alternative suggestions based on legislation and item databases of different states at will.
  • Another embodiment is capable of retrieving matching tariffs, used for comparatively suggesting a unique identification scheme for the end user to consider.
  • the disclosed invention proposes a method of allocating unique identification numbers, such as ones enforced through the Harmonized System (HS), to items and merchandise of any nature or area of use, with the aid of an artificial neural network chain capable of implicitly conceptualizing data within categories, chapters, positions and subpositions through dedicated neural networks trained for the suggestion of which; comprising modules and processing units undertaking said classification and allocation practices. Different modules interact with one another according to different embodiments; entirety of which are optionally and cooperatively employable.
  • a method for allocating unique identification numbers to items and merchandise in a systematic and conceptualized manner is proposed.
  • said unique identification number allocation method includes designing at least one artificial neural network with a fully-connected hidden layer.
  • said unique identification number allocation method includes training said artificial neural network with files in the form of training sets and producing models resulting therefrom.
  • said unique identification number allocation method includes forming a significance-based hierarchical sequence of said artificial neural networks i.e. output of one of said neural networks is the input of the next.
  • said unique identification number allocation method includes introducing a preprocessed text vector as input to the first level of said artificial neural network sequence, and.
  • said unique identification number allocation method includes established communication with at least one external module for accepting raw text input. In a further aspect of the present invention, said unique identification number allocation method includes a preprocessing step to produce input-compatible vectors from a raw text query.
  • said unique identification number allocation method includes lowercasing, where every character in the raw text query is converted to lowercase and recurrent words are eliminated.
  • said unique identification number allocation method includes parsing, where words are parsed and converted to UTF8 format, punctuation and non-characters are tossed, words are broken down to syllables and morphemes, and arranged in groups of two and three, forming tokens.
  • said unique identification number allocation method includes vector conversion, where tokenized and untokenized instances of data in the previous steps are converted to numeric vectors.
  • said unique identification number allocation method includes normalization, where vector(s) formed thus far are normalized in order to be confined to a specific interval.
  • said unique identification number allocation method includes said artificial neural network training step further comprising one first-level training scheme for categories; one second-level training scheme for chapters; one third-level training scheme for positions; one fourth-level training scheme for subpositions, and; one fifth-level training scheme for identification number(s) as a whole.
  • said unique identification number allocation method includes said first, second, third and fourth- level training schemes for categories, chapters, positions and subpositions respectively comprising divisions of digit pairs, starting from the leftmost and descending in the order of significance.
  • said unique identification number allocation method includes said message verification step authentication of said packet is having been acknowledging only if all steps are positively executed, conversely which it is tossed.
  • a system for allocating unique identification numbers to items and merchandise comprising at least one processing unit and one database is proposed.
  • said processing unit includes modules capable of executing parallelized instances of artificial neural networks.
  • said database includes lists pertaining to an item catalogue and a list of unique identification numbers associated therewith.
  • said database and processing unit include connection with additional modules selected from a group containing; legislative comparison module, similar item comparison module, matching tariffs module.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Software Systems (AREA)
  • Educational Administration (AREA)
  • Databases & Information Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Machine Translation (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

La présente invention concerne de façon générale un procédé destiné à attribuer des numéros spécifiques/uniques à tout article suggéré conformément, mais sans s'y limiter, au système harmonisé (HS) qui est la norme internationalement valide de nomenclature généralisée de marchandises, et un système mettant en œuvre ledit procédé pour désigner lesdits numéros spécifiques/uniques vers un article ou une marchandise suggérée, c'est-à-dire une approche orientée vers l'apprentissage automatique capable de traiter des textes donnés de manière conceptuelle pour produire une ou des prédictions exactes et précises.
EP18849466.0A 2018-07-04 2018-07-04 Procédé utilisant des réseaux de neurones artificiels pour trouver un code de système harmonisé unique à partir de textes donnés et système pour le mettre en ?uvre Pending EP3818478A1 (fr)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/TR2018/050344 WO2020009670A1 (fr) 2018-07-04 2018-07-04 Procédé utilisant des réseaux de neurones artificiels pour trouver un code de système harmonisé unique à partir de textes donnés et système pour le mettre en œuvre

Publications (1)

Publication Number Publication Date
EP3818478A1 true EP3818478A1 (fr) 2021-05-12

Family

ID=65520368

Family Applications (1)

Application Number Title Priority Date Filing Date
EP18849466.0A Pending EP3818478A1 (fr) 2018-07-04 2018-07-04 Procédé utilisant des réseaux de neurones artificiels pour trouver un code de système harmonisé unique à partir de textes donnés et système pour le mettre en ?uvre

Country Status (5)

Country Link
US (1) US20210312470A1 (fr)
EP (1) EP3818478A1 (fr)
CN (1) CN112513901A (fr)
IL (1) IL279828A (fr)
WO (1) WO2020009670A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11475493B2 (en) 2019-12-11 2022-10-18 Ul Llc Methods for dynamically assessing applicability of product regulation updates to product profiles
WO2021179138A1 (fr) * 2020-03-09 2021-09-16 图灵通诺（北京）科技有限公司 Procédé et système permettant d'analyser des produits sur une étagère de supermarché

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6542888B2 (en) * 1997-11-26 2003-04-01 International Business Machines Corporation Content filtering for electronic documents generated in multiple foreign languages
US7840986B2 (en) * 1999-12-21 2010-11-23 Tivo Inc. Intelligent system and methods of recommending media content items based on user preferences
US7957996B2 (en) 2004-03-31 2011-06-07 International Business Machines Corporation Market expansion through optimized resource placement
US7603349B1 (en) * 2004-07-29 2009-10-13 Yahoo! Inc. User interfaces for search systems using in-line contextual queries
US20130325770A1 (en) * 2012-06-05 2013-12-05 Sap Ag Probabilistic language model in contextual network
US8965820B2 (en) 2012-09-04 2015-02-24 Sap Se Multivariate transaction classification
KR101571041B1 (ko) 2013-12-10 2015-11-23 주식회사 한국무역정보통신 Hs 품목 분류 코드 결정 시스템
WO2016018488A2 (fr) * 2014-05-09 2016-02-04 Eyefluence, Inc. Systèmes et procédés de discernement de signaux oculaires et d'identification biométrique continue
WO2016057000A1 (fr) 2014-10-08 2016-04-14 Crimsonlogic Pte Ltd Classification de codes de tarifs douaniers
KR101596353B1 (ko) 2015-03-16 2016-02-22 주식회사 한국무역정보통신 품목분류코드 결정 장치 및 방법
RU2703343C2 (ru) * 2015-03-20 2019-10-16 Фраунхофер-Гезелльшафт Цур Фердерунг Дер Ангевандтен Форшунг Е.Ф. Назначение оценки релевантности для искусственных нейронных сетей
US11514096B2 (en) * 2015-09-01 2022-11-29 Panjiva, Inc. Natural language processing for entity resolution
EP3577570A4 (fr) * 2017-01-31 2020-12-02 Mocsy Inc. Extraction d'informations à partir de documents

Also Published As

Publication number Publication date
WO2020009670A1 (fr) 2020-01-09
US20210312470A1 (en) 2021-10-07
CN112513901A (zh) 2021-03-16
IL279828A (en) 2021-03-01

Similar Documents

Publication Publication Date Title
CN110968699B (zh) 一种基于事理推荐的逻辑图谱构建及预警方法和装置
CN110427623A (zh) 半结构化文档知识抽取方法、装置、电子设备及存储介质
CN109446341A (zh) 知识图谱的构建方法及装置
CN109598517B (zh) 商品通关处理、对象的处理及其类别预测方法和装置
CN105117426B (zh) 一种海关编码智能搜索方法
CN110069623A (zh) 摘要文本生成方法、装置、存储介质和计算机设备
CN110458324B (zh) 风险概率的计算方法、装置和计算机设备
CN112434024B (zh) 面向关系型数据库的数据字典生成方法、装置、设备及介质
CN115547466B (zh) 基于大数据的医疗机构登记评审系统及其方法
WO2020009670A1 (fr) Procédé utilisant des réseaux de neurones artificiels pour trouver un code de système harmonisé unique à partir de textes donnés et système pour le mettre en œuvre
CN116861269A (zh) 工程领域的多源异构数据融合及分析方法
KR20220126468A (ko) 우려거래자 정보 수집 및 관리 시스템과 그 방법
CN117151222A (zh) 领域知识引导的突发事件案例实体属性及其关系抽取方法、电子设备和存储介质
CN114429384B (zh) 基于电商平台的产品智能推荐方法及系统
Gao et al. Ensemble‐Based Neighborhood Attribute Reduction: A Multigranularity View
CN115309705A (zh) 一种自动识别城市信息模型平台基础数据元素的数据集成分类系统及其分类方法
US11836612B2 (en) Maintaining master data using hierarchical classification
CN114328813A (zh) 词标准化方法、装置、设备及存储介质
CN113947061A (zh) 一种基于多模组文本智能编码算法的海关进出口商品归类方法
CN113517045A (zh) 一种基于路径生成的电子病历icd代码预测方法及预测系统
CN111563178A (zh) 规则逻辑图对比方法、装置、介质及电子设备
CN112132534B (zh) 一种武器装备全寿命周期综合保障数据的管理方法及系统
Sonje et al. draw2code: Ai based auto web page generation from hand-drawn page mock-up
JP6886935B2 (ja) データ分析支援システムおよびデータ分析支援方法
CN117056458B (zh) 基于向量空间算法进行前端检索的方法

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20210122

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20220722